deepseek
DeepSeek: DeepSeek V3.2
deepseek/deepseek-v3.2
DeepSeek-V3.2 is a large language model designed to harmonize high computational efficiency with strong reasoning and agentic tool-use performance. It introduces DeepSeek Sparse Attention (DSA), a fine-grained sparse attention mechanism...
- Type
- text
- Provider
- deepseek
- Model ID
- deepseek/deepseek-v3.2
- Capabilities
- tools, reasoning
- Context window
- 164K tokens
- Self-hostable
- Yes — runs on your own GPU
- Input price
- $0.31 / 1M tokens
- Output price
- $0.46 / 1M tokens
Cloud price is billed from prepaid credits when a request fails over to the cloud. Open-weight models run free on GPUs you own — the gateway routes to your nodes first.
What it costs to run
cloud · 1M in + 1M out
$0.77
$0.31 for a million input tokens plus $0.46 for a million output tokens, billed from credits only when a request fails over to the cloud.
your hardware
$0 / token
Served from a workstation or on-prem server you own, there is no per-token fee — the marginal cost is power.
Run it on your own hardware
Estimated GPU memory for DeepSeek: DeepSeek V3.2 (685B parameters), including the KV cache for a 8,192-token context.
- Q4_K_M
- ~390.7 GBmulti-GPU or large unified memory
- Q8_0
- ~677.9 GBmulti-GPU or large unified memory
- FP16 / BF16
- ~1277.9 GBmulti-GPU or large unified memory
Install the agent on the machine and it dials out to the gateway over a single WebSocket — no port forwarding or inbound firewall rules. Requests for this model go to your node first and fail over to the cloud only when it can't serve.
/// initialize
Call DeepSeek: DeepSeek V3.2 through one OpenAI-compatible endpoint.
Serve it from hardware you own, with cloud failover when your nodes are busy or offline.
no credit card · 2 nodes free · openai-compatible
Frequently asked questions
- What is the context window of DeepSeek: DeepSeek V3.2?
- DeepSeek: DeepSeek V3.2 accepts up to 163,840 tokens (164K) of context per request.
- How much does DeepSeek: DeepSeek V3.2 cost per million tokens?
- Through Wide Area Intelligence, DeepSeek: DeepSeek V3.2 costs $0.31 per 1M input tokens and $0.46 per 1M output tokens when served from the cloud, billed from prepaid credits. A workload of 1M input plus 1M output tokens costs $0.77.
- Can I run DeepSeek: DeepSeek V3.2 on my own hardware?
- Yes — DeepSeek: DeepSeek V3.2 is an open-weight model you can run on a workstation or on-prem server you own. At Q4_K_M it needs roughly 390.7 GB of GPU memory with an 8,192-token context — more than a single 80 GB card, so plan for multiple GPUs or a large unified-memory machine. Wide Area Intelligence routes requests to your own nodes first and fails over to the cloud only when they can't serve.
More from deepseek
- DeepSeek: DeepSeek V3deepseek/deepseek-chat
- DeepSeek: DeepSeek V3 0324deepseek/deepseek-chat-v3-0324
- DeepSeek: DeepSeek V3.1deepseek/deepseek-chat-v3.1
- DeepSeek: DeepSeek V3.1 Terminusdeepseek/deepseek-v3.1-terminus
- DeepSeek: DeepSeek V3.2 Expdeepseek/deepseek-v3.2-exp
- DeepSeek: DeepSeek V4 Flash 0423deepseek/deepseek-v4-flash
- DeepSeek: DeepSeek V4 Flash 0731deepseek/deepseek-v4-flash-0731
- DeepSeek: DeepSeek V4 Flash Vision Expdeepseek/deepseek-v4-flash-vision-exp